Methods and related systems for detecting miscalibration of extrinsic lidar parameters

Abstract

Systems and methods for detecting extrinsic parameter miscalibration for light detection and ranging (“LiDAR”) devices are disclosed. According to one embodiment, a LiDAR detection method includes scanning, by a LiDAR system comprising one or more LiDAR devices, a field-of-view during one or more time periods, wherein the LiDAR system moves along a non-linear path during at least a subset of the one or more time periods. Return signal data obtained during at least the subset is aggregated and fiducial markers represented by the aggregated return signal data are identified. Each of the identified fiducial markers are compared to a respective reference fiducial marker of a plurality of reference fiducial markers. Based on the comparison, miscalibration of an extrinsic parameter associated with a position and/or an orientation of at least one of the LiDAR devices is detected. An action is initiated to remediate the detected miscalibration of the extrinsic parameter.

What is claimed is: 
     
         1 . A light detection and ranging (LiDAR) detection method comprising:
 scanning, by a LiDAR system comprising one or more LiDAR devices, a field-of-view during one or more time periods, wherein the LiDAR system moves along a non-linear path during at least a subset of the one or more time periods;   aggregating return signal data obtained during at least the subset of the one or more time periods;   identifying one or more fiducial markers represented by the aggregated return signal data;   comparing each of the identified fiducial markers to a respective reference fiducial marker of a plurality of reference fiducial markers;   detecting, based on the comparison, miscalibration of an extrinsic parameter associated with a position and/or an orientation of at least one of the one or more LiDAR devices; and   initiating an action to remediate the detected miscalibration of the extrinsic parameter.   
     
     
         2 . The method of  claim 1 , wherein scanning, by the LiDAR system comprising the one or more LiDAR devices, the field-of-view during the one or more time periods comprises:
 emitting, by one or more transmitters, one or more optical signals; and   receiving, by one or more receivers, one or more return signals corresponding to the one or more optical signals, wherein each of the one or more LiDAR devices includes at least one of the one or more transmitters and at least one of the one or more receivers.   
     
     
         3 . The method of  claim 1 , wherein the return signal data are derived from the one or more return signals and comprise a point cloud of measurements corresponding to the one or more optical signals. 
     
     
         4 . The method of  claim 3 , wherein the return signal data comprise one or more measurements of range from the LiDAR system to one or more objects in an environment of the LiDAR system. 
     
     
         5 . The method of  claim 1 , wherein the LiDAR system is mounted on or carried by a kinematic apparatus. 
     
     
         6 . The method of  claim 5 , further comprising:
 receiving, from the kinematic apparatus, directional data corresponding to respective paths of the LiDAR system during each of the one or more time periods.   
     
     
         7 . The method of  claim 6 , wherein aggregating the return signal data obtained during at least the subset of the one or more time periods comprises:
 aggregating, based on the directional data, a threshold amount of the return signal data obtained during the subset of the one or more time periods.   
     
     
         8 . The method of  claim 7 , wherein the directional data indicate the LiDAR system as changing direction within a range of turning radiuses or within a threshold turning radius during at least the subset of the one or more time periods. 
     
     
         9 . The method of  claim 1 , wherein the LiDAR system includes and/or accesses the plurality of reference fiducial markers and wherein the plurality of reference fiducial markers comprise data representing visual attributes of objects of standardized shapes and/or sizes. 
     
     
         10 . The method of  claim 9 , wherein the objects of standardized shapes and/or sizes comprise lane markings and/or street signs. 
     
     
         11 . The method of  claim 1 , wherein comparing each of the identified fiducial markers to the respective reference fiducial marker comprises:
 for each of the identified fiducial markers, selecting the respective reference fiducial marker based on data representing visual attributes of the respective identified fiducial marker.   
     
     
         12 . The method of  claim 1 , wherein comparing each of the identified fiducial markers to the respective reference fiducial marker comprises:
 identifying a distortion based on a difference between data representing one or more visual attributes of the identified fiducial marker and data representing one or more visual attributes of the respective reference fiducial marker; and   comparing the distortion to a distortion threshold.   
     
     
         13 . The method of  claim 12 , wherein initiating the action to remediate the detected miscalibration of the extrinsic parameter comprises:
 generating an alert recommending recalibration of one or more extrinsic parameters of the LiDAR system.   
     
     
         14 . The method of  claim 12 , wherein the distortion comprises a difference between a length, a width, and/or a height of the identified fiducial marker and the respective reference fiducial marker or a ratio between the length, the width, and/or the height of the identified fiducial marker and the respective reference fiducial marker. 
     
     
         15 . The method of  claim 12 , wherein a type of the distortion corresponds to whether the distortion is a difference between a length, a width, and/or a height of the identified fiducial marker and the respective reference fiducial marker. 
     
     
         16 . The method of  claim 12 , wherein the distortion comprises an average distortion for a subset of the one or more identified fiducial markers, wherein each identified fiducial marker in the subset of the one or more identified fiducial markers corresponds to a particular reference fiducial marker. 
     
     
         17 . A LiDAR system comprising:
 one or more LiDAR devices configured to scan a field-of-view during one or more time periods, wherein the LiDAR system moves along a non-linear path during at least a subset of the one or more time periods; and   a processing device configured to:
 aggregate return signal data obtained during at least the subset of the one or more time periods; 
 identify one or more fiducial markers represented by the aggregated return signal data; 
 compare each of the identified fiducial markers to a respective reference fiducial marker of a plurality of reference fiducial markers; 
 detect, based on the comparison, miscalibration of an extrinsic parameter associated with a position and/or an orientation of at least one of the one or more LiDAR devices; and 
 initiate an action to remediate the detected miscalibration of the extrinsic parameter. 
   
     
     
         18 . The system of  claim 17 , wherein the one or more LiDAR devices are configured to scan a field-of-view during one or more time periods by:
 emitting, by one or more transmitters, one or more optical signals; and   receiving, by one or more receivers, one or more return signals corresponding to the one or more optical signals, wherein each of the one or more LiDAR devices includes at least one of the one or more transmitters and at least one of the one or more receivers.   
     
     
         19 . The system of  claim 17 , wherein the return signal data are derived from the one or more return signals and comprise a point cloud of measurements corresponding to the one or more optical signals. 
     
     
         20 . The system of  claim 19 , wherein the return signal data comprise one or more measurements of range from the LiDAR system to one or more objects in an environment of the LiDAR system. 
     
     
         21 . The system of  claim 17 , wherein the LiDAR system is mounted on or carried by a kinematic apparatus. 
     
     
         22 . The system of  claim 21 , wherein the processing device is configured to:
 receive, from the kinematic apparatus, directional data corresponding to respective paths of the LiDAR system during each of the one or more time periods.   
     
     
         23 . The system of  claim 22 , wherein the processing device is configured to aggregate the return signal data obtained during at least the subset of the one or more time periods by:
 aggregating, based on the directional data, a threshold amount of the return signal data obtained during the subset of the one or more time periods.   
     
     
         24 . The system of  claim 23 , wherein the directional data indicate the LiDAR system as changing direction within a range of turning radiuses or within a threshold turning radius during at least the subset of the one or more time periods. 
     
     
         25 . The system of  claim 17 , wherein the processing device is configured to access the plurality of reference fiducial markers and wherein the plurality of reference fiducial markers comprise data representing visual attributes of objects of standardized shapes and/or sizes. 
     
     
         26 . The system of  claim 25 , wherein the objects of standardized shapes and/or sizes comprise lane markings and/or street signs. 
     
     
         27 . The system of  claim 17 , wherein the processing device is configured to compare each of the identified fiducial markers to the respective reference fiducial marker of the plurality of reference fiducial markers by:
 selecting, for each of the identified fiducial markers, the respective reference fiducial marker based on data representing visual attributes of the respective identified fiducial marker.   
     
     
         28 . The system of  claim 17 , wherein the processing device is configured to compare each of the identified fiducial markers to the respective reference fiducial marker of the plurality of reference fiducial markers by:
 identifying a distortion based on a difference between data representing one or more visual attributes of the identified fiducial marker and data representing one or more visual attributes of the respective reference fiducial marker; and   comparing the distortion to a distortion threshold.   
     
     
         29 . The system of  claim 28 , wherein the processing device is configured to initiate the action to remediate the detected miscalibration of the extrinsic parameter by:
 generating an alert recommending recalibration of one or more extrinsic parameters of the LiDAR system.   
     
     
         30 . The system of  claim 28 , wherein the distortion comprises a difference between a length, a width, and/or a height of the identified fiducial marker and the respective reference fiducial marker or a ratio between the length, the width, and/or the height of the identified fiducial marker and the respective reference fiducial marker. 
     
     
         31 . The system of  claim 28 , wherein a type of the distortion corresponds to whether the distortion is a difference between a length, a width, and/or a height of the identified fiducial marker and the respective reference fiducial marker. 
     
     
         32 . The system of  claim 28 , wherein the distortion comprises an average distortion for a subset of the one or more identified fiducial markers, wherein each identified fiducial marker in the subset of the one or more identified fiducial markers corresponds to a particular reference fiducial marker.